Silicon, next to oxygen, is the most prevalent element on earth, and is the most abundant mineral in the earth's crust. It occurs in nature as silica (silicon dioxide--SiO.sub.2) or corresponding silicic acids. Silicon is present in plants, and is widespread in foodstuffs, particularly monocotyledons such as grain, in clay (aluminum silicate), in sand, and in glass. In medicine, silicon is used therapeutically as magnesium trisilicate, and as organic compounds used as defoaming agents. Silicones (organosiloxanes) are used in various cosmetic surgical implant procedures. Due most probably to dietary intake, at least small amounts of silicon may be found in most animal tissues and fluids (Scot. Med. J. 27:17-19 (1982)).
Silicon is a trace element, comprising less than 0.01% of the human body. Silicon has been demonstrated as an essential element, i.e., one that is required for maintenance of life, and when deficient consistently results in an impairment of a function from optimal to suboptimal (Science 213:1332 (1981)).
Proof of the essentiality of silicon was independently established by two investigators. Carlisle established a silicon deficiency state incompatible with normal growth in chicks (Science 178:619 (1972)), and Schwartz and Milne showed similar results with rats (Nature, 239:33 (1972)). Using comparable methods, both studies showed that the animals responded to supplementation with sodium metasilicate with a 30 to 50 percent stimulation in growth. Subsequent examination of the animals raised on silicon-deficient diets revealed depressed bone growth and severe bone deformities, particularly of the skull.
Although silicon has been known as a regular constituent of biological materials since the beginning of the century, little is known about its metabolism. It is known that silicon specifically concentrates in the mitochondria of osteoblasts, and that it plays a role in bone and cartilage formation (Science 213:1332 (1981)). In addition, since silicon is present in high concentrations in collagen, it has been suggested that silicon plays a role in cross-linking connective tissues at the level of mucopolysaccharides (Fed. Proc. 33:1748 (1974)). It has therefore been postulated that apart from bone formation, silicon participates in growth and maintenance of connective tissue, as in embryonic development and wound healing, and in regulation of ions, metabolites and water in connective tissue. (Fed. Proc. 33:1758-1766 (1974)).
Silicon in foods and beverages is readily absorbed across the intestinal wall. Studies have shown that there is a narrow range of silicon concentration in the serum of healthy adults, and with the exception of urine, the concentrations of silicon in all other body fluids is similar to that of normal serum. Higher and wider ranges of silicon levels in the urine show that the kidney is the main excretory organ for silicon absorbed from the alimentary canal (Scot. Med. J. 27:17-19 (1982)).
The level of silicon in the blood and tissues has been shown to be affected by age, as well as by sex, castration, adrenalectomy and thyroidectomy (Ann. Endocrinol. 32:397 (1971)). The silicon content of the aorta, skin and thymus in the rabbit, rat, chicken and pig was found to significantly decline with age, whereas other tissues such as the heart, kidney, muscle and tendon show little or no change (Fed. Proc. 33:1758-1766 (1974)). In addition, the silicon content of the dermis of human skin has been shown to diminish with age (J. Biol. Chem. 75:789-794 (1927)). In contrast, Leslie et al. showed an increase in rat brain, liver, spleen, lung and femur silicon with age (Proc. Soc. Exptl. Bio. Med. 110:218 (1962)). And Kworning et al. described elevated silicon deposition in the human aorta wall during aging (J. Geront. 5:23-25 (1960)). In addition, it has been demonstreated that silicon was elevated in the aorta with focal atherosclerosis, as well as in the atherosclerotic focus itself (Folia Morph. 25:353-356 (1977)). Further, it has been reported that with advancing age, the SiO.sub.2 level of human peribronchial lymph glands gradually increases even in those who have no history of exposure to dust (J. Pathol. 51:269-275 (1940)). Our own work, moreover, has demonstrated an increase of kidney silicon levels in normal rats with aging.
Although silicon is an essential trace element for human growth and is necessary for bone formation, silicon intoxication has been shown to cause various diseases. In addition to cases of acute toxicity, there is justifiable suspicion that the pathogenesis of some chronic diseases may be related to prolonged exposure to concentrations of toxins insufficient to produce conspicuous manifestations (J. Chron. Dis. 27:135-161 (1974)). For example, a substantial portion of patients with terminal renal failure have no clearly definable etiology of their renal disease. It may be speculated therefore that some renal disease may be associated with chronic exposure to certain toxins, including silicon.
Much information is known about the toxic effects of silicon in the lung. Varying amounts of silica normally enter the respiratory tract across the lung barrier as silicic acid and are eventually eliminated. Prolonged inhalation and accumulation of fine particulate silica in the lung, however, produces a pulmonary inflammatory response, granuloma formation and chronic fibrosis (silicosis) (Prin. Int. Med., 9th Ed., Isselbacher et al. (eds), McGraw-Hill Book Co., N.Y. 1980). In silicosis, the injury seems to be related to both the crystal structure of the silicon and the host response. Workers in stone quarries, or in other industries where sand or other silicate dusts are prevalent, are prone to contract this disease.
It is commonly believed that ingested silicates are both inert and nonabsorbable, but there has long been a suspicion that silicates are nephrotoxic in humans (Scot. Med. J. 27:10-17 (1982)). In 1922, Gye and Purdy investigated the toxicity of parenterally administered colloidal silica in rabbits which resulted in interstitial nephritis, hepatic fibrosis and splenomegaly within a period of weeks to several months (Br. J. Exp. Path. 3:75-85 (1922)). These findings were later confirmed by Schepers et al. (AMA Arch. Industr. Hlth. 15:599 (1957)). In 1970, Newberne and Wilson showed that oral administration of certain silicates produced significant renal tubular damage and chronic interstitial inflammation in dogs (Proc. Nat. Acad. Sci. 65:872-875 (1970)). And in 1982, Dobbie and Smith showed that oral ingestion of magnesium trisilicate resulted in renal damage in guinea pigs in four months (Scot. Med. J. 27:10 (1982)). Recently, Shainkin-Kestenbaum et al. showed in in vivo studies that silicon in concentrations similar to those found in the serum of uremic patients inhibited superoxide dismutase, an enzyme that destroys oxygen free radicals which can damage cells (J. Trace Elem. Electrolytes Health Dis. 4:97 (1990)).
In humans, chronic exposure to silica has been associated with mild renal functional abnormalities and minor histologic changes in the kidneys. Bolton et al. reported four patients with a history of intense silica exposure and rapidly progressive renal failure, and concluded that silicon appeared to be responsible for the nephrotoxic changes (Am. J. Med. 71:823 (1981)). Silicon has also been shown to have a direct dose-dependent toxic effect on the kidney (J. Pathol. 103:35-40 (1970)), and silicon particles are cytotoxic, as shown by studies demonstrating damage to macrophages ingesting silicon (Am. Rev. Respir. Dis. 113:643-665 (1976)).
Since it is known that the principal organ of silicon elimination is the kidney, it is not surprising that an increase in plasma silicon levels (Biomedicine 33:228-230 (1980)), as well as an increase in certain tissue silicon levels, have been reported in studies of patients suffering from chronic renal failure and in patients on hemodialysis. (J. Chron. Dis. 27:135-161 (1974)). The accumulation of increased quantities of silicon in renal failure results from its decreased renal clearance (J. Chron. Dis. 27:135-161 (1974)). The high serum silicon levels demonstrated in hemodialysis patients have been associated with osteitis fibrosa (Xth Intl. Cong. of Nephr., Jun. 26-31, 1987), and elevated cerebral spinal fluid (CSF) silicon levels have been observed in patients with chronic renal insufficiency where CSF silicon levels increased as renal function declined (Neurology: 86-789 (1983)). It has been hypothesized, therefore, that since silicon is nephrotoxic and accumulates in blood and body tissues of patients with renal failure, silicon may contribute to the steady progression of renal failure once initiated. (Id.)
In addition to silicon, aluminum has been found to accumulate in advanced kidney disease patients on chronic hemodialysis. Currently, the most effective means of increased removal of aluminum during hemodialysis is by chelation with desferrioxamine (DFO) (Clin. Nephr. 24:594-597 (1985)). At the end of a dialysis treatment, the chelator is administered to the patient, whereupon at the next dialysis session, the aluminum-DFO complex is removed. Various dialysis related modalities may be used to remove the aluminum-DFO complex including hemodialysis, peritoneal dialysis, hemofiltration or charcoal (or resin) hemoperfusion (Kid. Int.times.l 33 suppl. 24:5-171 (1988)). Known side effects of DFO treatment include anaphylactic reactions, abdominal pain, posterior cataracts, visual impairments and predisposition to development of fungal infections. DFO has not yet been investigated for its ability to form stable complexes with silicon (Clin. Neph. 24 at Table 1, p. 595). A need continues to exist, therefore, for a chelator that would help promote the removal of silicon accumulation in patients with advanced kidney disease on chronic hemodialysis.
Impaired renal function can cause secondary hypertension. (Clin. and Exp. Hypertension 3(4):775 (1981)). Thus, any treatment, including the reduction of silicon, which improves renal function may also reduce some forms of hypertension.
The cause of essential hypertension is unclear, but several laboratories have detected a low molecular weight protein (MW=12,000-15,000) in the plasma of many patients with essential hypertension and in animals with hereditary forms of hypertension (Hypertension 5:128 (1983); Electrophoresis 6:292 (1985); J. Hypertension 3(3):249 (1985); Biochem. Biophys. Res. Commun. 111(3):1015 (1983)). This protein has been called the "hypertension associated protein", or HAP.
Our laboratory has partially purified HAP from plasma of a patient with volume-expansion related hypertension (primary aldosteronism). The semi-purified HAP was shown to have the properties attributable to natriuretic hormone, i.e., to induce natriuresis in a test animal, to inhibit the active transport enzyme, Na-K-ATPase, and to displace .sup.3 H-ouabain from its binding sites on purified Na-K-ATPase (Weiler, et al., "Characterization of a Volume-Responsive Low Molecular Weight Protein in Human Plasma"). We have also demonstrated that a low molecular weight Na-K-ATPase inhibitor dissociated from HAP can independently induce vasoconstriction and potentiate the vasoconstriction caused by norepinephrine in isolated blood vessels (Gonick, et al., "Comparison of Low Molecular Weight Plasma and Urine Na-K-ATPase Inhibitors/Hypertensive Factors").
Hypertension can occur when the blood vessels constrict. Therefore, one way to reduce blood pressure is to dilate the blood vessels. A hormonal system locally produced in blood vessels is one system which controls dilation and constriction of blood vessels. Endothelium-derived relaxing factor (EDRF), which has been equated to nitric oxide (NO) or closely related compounds, causes vasodilation by activating guanylate cyclase within the vascular smooth muscle. The guanylate cyclase produces cyclic GMP (cGMP), which decreases the intracellular concentration of calcium cations (Ca.sup.2+), resulting in vasodilation, decreased peripheral vascular resistance, and reduced blood pressure (Annu. Rev. Pharmacol. Toxicol. 30:535 (1990); American J. Hypertension 3:317 (1990)). Kondo et al. have found that superoxide dismutase increased the basal activity of guanylate cyclase, probably through the removal of superoxide anions (O.sub.2.sup.-) (Br. J. Pharmacol. 98:630 (1989)). Luscher et al. noted that superoxide anions destroy EDRF, while the free radial scavenger superoxide dismutase prolongs its half-life, and that the main disorder in atherosclerotic arteries is a decreased release of EDRF (Kidney Int'l 39:575 (1991)). Therefore, a compound stimulating EDRF or superoxide dismutase should cause a reduction in blood pressure.
Reactive oxygen species appear to play a role in the progression and exacerbation of diabetic nephropathy, and in the initiation and/or evolution of microangiopathy and vascular disease in diabetic mellitus patients. E.g., Watanabe, A., Nippon Jinzo Gakkai Shi, 34 (11):1219-25 (1992); Katoh, K., Diabetes Res. Clin. Pract., 18(2):89-98 (1992); Collier, et al. , Diabetes, 41(8):909-13 (1992). Accordingly, a compound which reduces reactive oxygen species should have a positive impact on these conditions.
Silicon may also be a neurotoxin. Silicon, together with aluminum, are significantly elevated in Alzheimer's disease in the neurofibrillary tangles, and in senile dementia there is a diffuse increase in silicon levels in the brain (Science 208:297-298 (1980)). Nikaido et al. demonstrated that patients with Alzheimer's disease showed a substantial increase of silicon in the cores and rims of the senile plaques. (Arch. Neurol. 27:549-554 (1922)).
Meso-2,3-dimercaptosuccinic acid (DMSA) is a water soluble compound analogous to 2,3-dimercaptopropanol (BAL). In contrast to BAL, however, DMSA is less toxic, has greater water solubility, limited lipid solubility, and is effective when given orally (Fund. Appl. Tox. 11:715-722 (1988)). DMSA may be administered orally or parenterally. A preferred dosage of DMSA for humans is 10-30 mg/kg daily.
DMSA is available as a white crystalline powder and exists in two forms, the meso form and the DL form. Because meso-DMSA is easier to synthesize and purify, it is more readily available and has been used in most published investigations. Meso-DMSA (m.p. 210.degree.-211.degree. C.) is sparingly soluble and must be titrated to approximately pH 5.5 to go into solution, or dissolved in 5% NaHCO.sub.3. The DL form (m.p. 124-125), on the other hand, is readily soluble in distilled water. (Ann. Rev. Pharmacol. Toxicol. 23:193-215 (1983)). As used herein, DMSA includes but is not limited to the meso, racemic and D and L isomers, whether derived from isomeric resolution of the racemic form or derived from stereospecific synthesis. DMSA is available from a variety of biochemical specialty firms.
DMSA was originally introduced by Friedheim and DaSilva in 1954 to promote uptake of antimony during schistosomiasis therapy (J. Pharm. Exp. Therap. 246:84 (1988)), and was first recognized as an antidote for heavy metal toxicity by Liang et al. in 1957 (Acta Physiol. Sin. 21:24-32 (1957)). Since then, DMSA has been shown to remove toxic forms of lead, mercury and arsenic from the body via urinary excretion, presumably by forming water-soluble metal complexes or chelates (Anal. Biochem. 160:217-226 (1987)).
DMSA has been shown to have variable success as an antidote for other toxicities. DMSA was reported to be effective at reducing the concentration of aluminum in the liver, spleen and kidney (Res. Com. Chem. Pathol. Pharm. 53:93-104 (1986)), reducing the concentration of cobalt in the liver, brain, heart and blood (Arch. Toxicol. 58:278-281 (1986), and as an antagonist for acute oral cadmium chloride intoxication by increasing the urinary elimination of cadmium (Tox Appl. Pharm. 66:361-367 (1982)). DMSA, however, did not increase urinary and fecal excretion of cobalt (Arch. Toxicol. 58:278-281 (1986)), and showed lower efficacy than other chelating agents as an antidote for zinc poisoning (Arch. Toxicol. 61:321-323 (1988)). (See Ann. Rev. Pharm. Toxicol. 23:193-215 (1983) for a review of the success and failure of DMSA in treating toxicities).
DMSA has also been labeled with .sup.99 Tc for use in renal scanning (J. Nucl. Med. 16:28-32 (1973), tumor detection (Clin. Otalary 12:405-411 (1987) and Clin. Nucl. Med. 13:159-165 (1988)), and for imaging myocardial infarcts (Clin. Nucl. Med. 12:514-518 (1987)).
DMSA has been reported as an effective and relatively nontoxic agent for treatment of metal poisoning. Other chelating agents have also been used as antidotes for metal toxicities, but these drugs have been shown to have many side effects. BAL is administered by a painful intramuscular injection and can cause nausea, vomiting and severe headache. Calcium disodium ethylenediaminetetraacetic acid (CaNa.sub.2 EDTA) must be administered parenterally, either intravenously or intramuscularly. It is painful when given intramuscularly, and when given in excessive dosage, can cause nephrotoxicity. Penicillamine is administered orally but is not as effective as BAL or CaNa.sub.2 EDTA. Additionally, it can cause reactions resembling penicillin sensitivity, is potentially nephrotoxic and causes neutropenia (Clinical Tox. 25:39-51 (1987).
To date, there are no known chelating agents effective for silicon removal, as well as no previously demonstrated effects of silicon removal. A need exists, therefore, for a method to remove silicon from the body, thereby improving blood pressure and kidney function, reducing neurological toxicities, and returning silicon to youthful levels.
The entire disclosures of the publications and references referred to above and hereafter in this specification are incorporated herein by reference.